From: Luis Chamberlain <mcgrof@kernel.org>
To: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>,
Christoph Hellwig <hch@infradead.org>,
"Eric W. Biederman" <ebiederm@xmission.com>,
ast@kernel.org, axboe@kernel.dk, bfields@fieldses.org,
bridge@lists.linux-foundation.org, chainsaw@gentoo.org,
christian.brauner@ubuntu.com, chuck.lever@oracle.com,
davem@davemloft.net, dhowells@redhat.com,
gregkh@linuxfoundation.org, jarkko.sakkinen@linux.intel.com,
jmorris@namei.org, josh@joshtriplett.org, keescook@chromium.org,
keyrings@vger.kernel.org, kuba@kernel.org,
lars.ellenberg@linbit.com, linux-fsdevel@vger.kernel.org,
linux-kernel@vger.kernel.org, linux-nfs@vger.kernel.org,
linux-security-module@vger.kernel.org,
nikolay@cumulusnetworks.com, philipp.reisner@linbit.com,
ravenexp@gmail.com, roopa@cumulusnetworks.com, serge@hallyn.com,
slyfox@gentoo.org, viro@zeniv.linux.org.uk,
yangtiezhu@loongson.cn, netdev@vger.kernel.org,
markward@linux.ibm.com, mcgrof@kernel.org,
linux-s390 <linux-s390@vger.kernel.org>
Subject: Re: linux-next: umh: fix processed error when UMH_WAIT_PROC is used seems to break linux bridge on s390x (bisected)
Date: Wed, 1 Jul 2020 15:38:59 +0000 [thread overview]
Message-ID: <20200701153859.GT4332@42.do-not-panic.com> (raw)
In-Reply-To: <8d714a23-bac4-7631-e5fc-f97c20a46083@i-love.sakura.ne.jp>
On Wed, Jul 01, 2020 at 11:08:57PM +0900, Tetsuo Handa wrote:
> On 2020/07/01 22:53, Luis Chamberlain wrote:
> >> Well, it is not br_stp_call_user() but br_stp_start() which is expecting
> >> to set sub_info->retval for both KWIFEXITED() case and KWIFSIGNALED() case.
> >> That is, sub_info->retval needs to carry raw value (i.e. without "umh: fix
> >> processed error when UMH_WAIT_PROC is used" will be the correct behavior).
> >
> > br_stp_start() doesn't check for the raw value, it just checks for err
> > or !err. So the patch, "umh: fix processed error when UMH_WAIT_PROC is
> > used" propagates the correct error now.
>
> No. If "/sbin/bridge-stp virbr0 start" terminated due to e.g. SIGSEGV
> (for example, by inserting "kill -SEGV $$" into right after "#!/bin/sh" line),
> br_stp_start() needs to select BR_KERNEL_STP path. We can't assume that
> /sbin/bridge-stp is always terminated by exit() syscall (and hence we can't
> ignore KWIFSIGNALED() case in call_usermodehelper_exec_sync()).
Ah, well that would be a different fix required, becuase again,
br_stp_start() does not untangle the correct error today really.
I also I think it would be odd odd that SIGSEGV or another signal
is what was terminating Christian's bridge stp call, but let's
find out!
Note we pass 0 to the options to wait so the mistake here could indeed
be that we did not need KWIFSIGNALED(). I was afraid of this prospect...
as it other implications.
It means we either *open code* all callers, or we handle this in a
unified way on the umh. And if we do handle this in a unified way, it
then begs the question as to *what* do we pass for the signals case and
continued case. Below we just pass the signal, and treat continued as
OK, but treating continued as OK would also be a *new* change as well.
For instance (this goes just boot tested, but Christian if you can
try this as well that would be appreciated):
diff --git a/include/linux/sched/task.h b/include/linux/sched/task.h
index bba06befbff5..d1898f5dd1fc 100644
--- a/include/linux/sched/task.h
+++ b/include/linux/sched/task.h
@@ -105,10 +105,12 @@ extern long kernel_wait4(pid_t, int __user *, int, struct rusage *);
/* Only add helpers for actual use cases in the kernel */
#define KWEXITSTATUS(status) (__KWEXITSTATUS(status))
+#define KWTERMSIG(status) (__KWTERMSIG(status))
+#define KWSTOPSIG(status) (__KWSTOPSIG(status))
#define KWIFEXITED(status) (__KWIFEXITED(status))
-
-/* Nonzero if STATUS indicates normal termination. */
-#define __KWIFEXITED(status) (__KWTERMSIG(status) == 0)
+#define KWIFSIGNALED(status) (__KWIFSIGNALED(status))
+#define KWIFSTOPPED(status) (__KWIFSTOPPED(status))
+#define KWIFCONTINUED(status) (__KWIFCONTINUED(status))
/* If KWIFEXITED(STATUS), the low-order 8 bits of the status. */
#define __KWEXITSTATUS(status) (((status) & 0xff00) >> 8)
@@ -116,6 +118,24 @@ extern long kernel_wait4(pid_t, int __user *, int, struct rusage *);
/* If KWIFSIGNALED(STATUS), the terminating signal. */
#define __KWTERMSIG(status) ((status) & 0x7f)
+/* If KWIFSTOPPED(STATUS), the signal that stopped the child. */
+#define __KWSTOPSIG(status) __KWEXITSTATUS(status)
+
+/* Nonzero if STATUS indicates normal termination. */
+#define __KWIFEXITED(status) (__KWTERMSIG(status) == 0)
+
+/* Nonzero if STATUS indicates termination by a signal. */
+#define __KWIFSIGNALED(status) \
+ (((signed char) (((status) & 0x7f) + 1) >> 1) > 0)
+
+/* Nonzero if STATUS indicates the child is stopped. */
+#define __KWIFSTOPPED(status) (((status) & 0xff) == 0x7f)
+
+/* Nonzero if STATUS indicates the child continued after a stop. */
+#define __KWIFCONTINUED(status) ((status) == __KW_CONTINUED)
+
+#define __KW_CONTINUED 0xffff
+
extern void free_task(struct task_struct *tsk);
/* sched_exec is called by processes performing an exec */
diff --git a/kernel/umh.c b/kernel/umh.c
index f81e8698e36e..c98fb1ed90c9 100644
--- a/kernel/umh.c
+++ b/kernel/umh.c
@@ -156,6 +156,18 @@ static void call_usermodehelper_exec_sync(struct subprocess_info *sub_info)
*/
if (KWIFEXITED(ret))
sub_info->retval = KWEXITSTATUS(ret);
+ /*
+ * Do we really want to be passing the signal, or do we pass
+ * a single error code for all cases?
+ */
+ else if (KWIFSIGNALED(ret))
+ sub_info->retval = KWTERMSIG(ret);
+ /* Same here */
+ else if (KWIFSTOPPED((ret)))
+ sub_info->retval = KWSTOPSIG(ret);
+ /* And are we really sure we want this? */
+ else if (KWIFCONTINUED((ret)))
+ sub_info->retval = 0;
}
/* Restore default kernel sig handler */
next prev parent reply other threads:[~2020-07-01 15:39 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-10 15:49 [PATCH 0/5] kmod/umh: a few fixes Luis R. Rodriguez
2020-06-10 15:49 ` [PATCH 1/5] selftests: kmod: Use variable NAME in kmod_test_0001() Luis R. Rodriguez
2020-06-10 15:49 ` [PATCH 2/5] kmod: Remove redundant "be an" in the comment Luis R. Rodriguez
2020-06-10 15:49 ` [PATCH 3/5] test_kmod: Avoid potential double free in trigger_config_run_type() Luis R. Rodriguez
2020-06-10 15:49 ` [PATCH 4/5] umh: fix processed error when UMH_WAIT_PROC is used Luis R. Rodriguez
2020-06-23 14:11 ` linux-next: umh: fix processed error when UMH_WAIT_PROC is used seems to break linux bridge on s390x (bisected) Christian Borntraeger
2020-06-23 14:23 ` Christian Borntraeger
2020-06-24 11:11 ` Christian Borntraeger
2020-06-24 12:05 ` Luis Chamberlain
2020-06-24 13:17 ` Luis Chamberlain
2020-06-24 16:13 ` Luis Chamberlain
2020-06-24 14:43 ` Christoph Hellwig
2020-06-24 15:54 ` Christian Borntraeger
2020-06-24 16:09 ` Luis Chamberlain
2020-06-24 17:58 ` Christian Borntraeger
2020-06-24 18:09 ` Christian Borntraeger
2020-06-24 18:32 ` Christian Borntraeger
2020-06-24 18:37 ` Christian Borntraeger
2020-06-25 13:26 ` Christian Borntraeger
2020-06-26 2:54 ` Luis Chamberlain
2020-06-26 5:22 ` Christian Borntraeger
2020-06-26 9:00 ` Christoph Hellwig
2020-06-26 11:40 ` Luis Chamberlain
2020-06-26 11:50 ` Luis Chamberlain
2020-06-30 17:57 ` Luis Chamberlain
2020-07-01 10:08 ` Christian Borntraeger
2020-07-01 13:24 ` Tetsuo Handa
2020-07-01 13:53 ` Luis Chamberlain
2020-07-01 14:08 ` Tetsuo Handa
2020-07-01 15:38 ` Luis Chamberlain [this message]
2020-07-01 15:48 ` Christian Borntraeger
2020-07-01 15:58 ` Luis Chamberlain
2020-07-01 16:01 ` Christian Borntraeger
2020-07-02 4:26 ` Tetsuo Handa
2020-07-02 19:46 ` Luis Chamberlain
2020-07-03 0:52 ` Tetsuo Handa
2020-07-03 13:28 ` Luis Chamberlain
2020-07-01 15:26 ` Christian Borntraeger
2020-07-01 13:46 ` Luis Chamberlain
2020-06-10 15:49 ` [PATCH 5/5] selftests: simplify kmod failure value Luis R. Rodriguez
2020-06-18 0:43 ` [PATCH 0/5] kmod/umh: a few fixes Andrew Morton
2020-06-19 20:46 ` Luis Chamberlain
2020-06-19 21:07 ` Luis Chamberlain
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200701153859.GT4332@42.do-not-panic.com \
--to=mcgrof@kernel.org \
--cc=ast@kernel.org \
--cc=axboe@kernel.dk \
--cc=bfields@fieldses.org \
--cc=borntraeger@de.ibm.com \
--cc=bridge@lists.linux-foundation.org \
--cc=chainsaw@gentoo.org \
--cc=christian.brauner@ubuntu.com \
--cc=chuck.lever@oracle.com \
--cc=davem@davemloft.net \
--cc=dhowells@redhat.com \
--cc=ebiederm@xmission.com \
--cc=gregkh@linuxfoundation.org \
--cc=hch@infradead.org \
--cc=jarkko.sakkinen@linux.intel.com \
--cc=jmorris@namei.org \
--cc=josh@joshtriplett.org \
--cc=keescook@chromium.org \
--cc=keyrings@vger.kernel.org \
--cc=kuba@kernel.org \
--cc=lars.ellenberg@linbit.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nfs@vger.kernel.org \
--cc=linux-s390@vger.kernel.org \
--cc=linux-security-module@vger.kernel.org \
--cc=markward@linux.ibm.com \
--cc=netdev@vger.kernel.org \
--cc=nikolay@cumulusnetworks.com \
--cc=penguin-kernel@i-love.sakura.ne.jp \
--cc=philipp.reisner@linbit.com \
--cc=ravenexp@gmail.com \
--cc=roopa@cumulusnetworks.com \
--cc=serge@hallyn.com \
--cc=slyfox@gentoo.org \
--cc=viro@zeniv.linux.org.uk \
--cc=yangtiezhu@loongson.cn \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).